Address Resolution

Joel Halpern <jmh@petunia.network.com> Mon, 19 November 1990 19:39 UTC

Received: from [137.192.1.9] by NRI.NRI.Reston.VA.US id aa29390; 19 Nov 90 14:39 EST
Received: from petunia.network.com by nsco.network.com (5.61/1.34) id AA03212; Mon, 19 Nov 90 13:40:04 -0600
Received: by petunia.network.com (4.1/SMI-4.1) id AA00921; Mon, 19 Nov 90 13:41:18 CST
Date: Mon, 19 Nov 90 13:41:18 CST
From: Joel Halpern <jmh@petunia.network.com>
Message-Id: <9011191941.AA00921@petunia.network.com>
To: frame-relay@NRI.Reston.VA.US
Subject: Address Resolution

Whether one uses PPP (with LQM negotiated off), or not, there is
a question about address resolution.  Several people have suggested
that the multi-cast solutions are too expensive in bandwidth and
excessive bits transmitted.  This may well be true.  However,  I
would sincerely hope that we are not going to insist on the full
overhead of X.500 when we need to resolve an address?

I would suggest that a server aimed specifically at the lower layer
service, such as an ES-IS server as envisioned by ISO, might be a
more effective solution to this problem.  

I personally rather like the idea of a multi-cast address which such
servers will listen to.  Then a box will send to that address when it
starts up, and not the responses.  Then it can use those address for
address resolution requests.  This should simplify customer configuration.

Joel M. Halpern			jmh@nsco.network.com
Network Systems Corporation